隨著影像縮放(Image Scaling)，或稱為影像超解析度 (Super-Resolution, SR)技術的發展，它們不僅廣泛應用於自然影像與醫學影像中，也在遙測影像 (Remote Sensing Image, RSI)領域中扮演關鍵角色。
RSI於土地覆蓋分類、目標偵測、語意分割與變遷偵測應用中愈加普及，如何透過影像縮放或超解析度技術獲得高解析度 (High-Resolution, HR) 的RSI影像以提升偵測演算法之準確率，且同時保持低模型複雜度與運算成本並支援任意倍率以部署於衛星端或邊緣裝置進行即時解析度提升已成為一項研究課題。
為解決此問題，本論文提出了輕量級模型架構，在此架構的特徵擷取模組加入倍率感知蒸餾式卷積層模組，使模型於不同目標倍率能夠動態調整以提取對RSI的重要特徵。接著採用了LMLTE作為上採樣模組，以較少的參數量實現任意倍率的影像重建。
實驗結果顯示，在固定倍率與任意倍率指標評估中，所提出的方法能在參數量較少的情況下優於近年應用於RSI的相關方法，因此，此方法不只能夠支援以單一模型處理任意倍率，同時也適合用於邊緣裝置。

With the advancement of image scaling, also known as super-resolution (SR) technology, it has found widespread applications not only in natural and medical imaging but also plays a crucial role in the field of remote sensing imagery (RSI).
RSI has become increasingly prevalent in applications such as land cover classification, object detection, semantic segmentation, and change detection. A key research challenge is how to obtain high-resolution (HR) RSI through image scaling or super-resolution techniques to improve the accuracy of detection algorithms, while maintaining low model complexity and computational cost, and supporting arbitrary scaling factors for deployment on satellites or edge devices for real-time resolution enhancement.
To address this issue, this thesis proposes a lightweight model architecture. In the feature extraction module, a scale-aware distillation-based convolutional block is introduced, enabling the model to dynamically adapt to different target scales and extract critical features from RSI. The LMLTE module is employed for upsampling, achieving arbitrary-scale image reconstruction with fewer parameters.
Experimental results demonstrate that the proposed method outperforms related RSI methods from the past few years under both fixed-scale and arbitrary-scale evaluation metrics, despite having fewer parameters. Therefore, the proposed method not only supports arbitrary-scale processing with a single model but is also well-suited for deployment on edge devices.

[1] Dong, C., Loy, C. C., He, K., & Tang, X. (2014). Learning a deep convolutional network for image super-resolution. In Computer Vision–ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part IV 13 (pp. 184-199). Springer International Publishing.
[2] Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., ... & Adam, H. (2017). Mobilenets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704.04861.
[3] He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 770-778).
[4] Dong, C., Loy, C. C., & Tang, X. (2016). Accelerating the super-resolution convolutional neural network. In Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11-14, 2016, Proceedings, Part II 14 (pp. 391-407). Springer International Publishing.
[5] B. Lim, S. Son, H. Kim, S. Nah and K. M. Lee, "Enhanced Deep Residual Networks for Single Image Super-Resolution," 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Honolulu, HI, USA, 2017, pp. 1132-1140, doi: 10.1109/CVPRW.2017.151.
[6] J. Hu, L. Shen and G. Sun, "Squeeze-and-Excitation Networks," 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, 2018, pp. 7132-7141, doi: 10.1109/CVPR.2018.00745.
[7] Y. Zhang, Y. Tian, Y. Kong, B. Zhong and Y. Fu, "Residual Dense Network for Image Super-Resolution," 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, 2018, pp. 2472-2481, doi: 10.1109/CVPR.2018.00262.
[8] Yulun Zhang, Kunpeng Li, Kai Li, Lichen Wang, Bineng Zhong, and Yun Fu. 2018. Image Super-Resolution Using Very Deep Residual Channel Attention Networks. In Computer Vision – ECCV 2018: 15th European Conference, Munich, Germany, September 8–14, 2018, Proceedings, Part VII. Springer-Verlag, Berlin, Heidelberg, 294–310. https://doi.org/10.1007/978-3-030-01234-2_18
[9] Agustsson, E., & Timofte, R. (2017). Ntire 2017 challenge on single image super-resolution: Dataset and study. In Proceedings of the IEEE conference on computer vision and pattern recognition workshops (pp. 126-135).
[10] Y. Yang and S. Newsam, “Bag-of-visual-words and spatial extensions for land-use classification,” in Proc. 18th SIGSPATIAL Int. Conf. Adv. Geograph. Inf. Syst., Nov. 2010, pp. 270–279.
[11] X. Hu, H. Mu, X. Zhang, Z. Wang, T. Tan and J. Sun, "Meta-SR: A Magnification-Arbitrary Network for Super-Resolution," 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA, 2019, pp. 1575-1584, doi: 10.1109/CVPR.2019.00167.
[12] L. Wang, Y. Wang, Z. Lin, J. Yang, W. An and Y. Guo, "Learning A Single Network for Scale-Arbitrary Super-Resolution," 2021 IEEE/CVF International Conference on Computer Vision (ICCV), Montreal, QC, Canada, 2021, pp. 4781-4790, doi: 10.1109/ICCV48922.2021.00476.
[13] Y. Chen, S. Liu and X. Wang, "Learning Continuous Image Representation with Local Implicit Image Function," 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA, 2021, pp. 8624-8634, doi: 10.1109/CVPR46437.2021.00852.
[14] J. Lee and K. H. Jin, "Local Texture Estimator for Implicit Representation Function," 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, LA, USA, 2022, pp. 1919-1928, doi: 10.1109/CVPR52688.2022.00197.
[15] Z. He and Z. Jin, "Latent Modulated Function for Computational Optimal Continuous Image Representation," 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, 2024, pp. 26026-26035, doi: 10.1109/CVPR52733.2024.02459.
[16] X. Wang, X. Chen, B. Ni, H. Wang, Z. Tong and Y. Liu, "Deep Arbitrary-Scale Image Super-Resolution via Scale-Equivariance Pursuit," 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, BC, Canada, 2023, pp. 1786-1795, doi: 10.1109/CVPR52729.2023.00178.
[17] Z. Hui, X. Wang and X. Gao, "Fast and Accurate Single Image Super-Resolution via Information Distillation Network," 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, 2018, pp. 723-731, doi: 10.1109/CVPR.2018.00082.
[18] Zheng Hui, Xinbo Gao, Yunchu Yang, and Xiumei Wang. Lightweight image super-resolution with information multi distillation network. In Proceedings of the Acm International Conference on Multimedia, pages 2024–2032, 2019.
[19] Jie Liu, Jie Tang, and Gangshan Wu. Residual feature distillation network for lightweight image super-resolution. In Proceedings of the European Conference on Computer Vi sion Workshops, pages 41–55, 2020.
[20] J. Liu, W. Zhang, Y. Tang, J. Tang and G. Wu, "Residual Feature Aggregation Network for Image Super-Resolution," 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, 2020, pp. 2356-2365, doi: 10.1109/CVPR42600.2020.00243.
[21] Z. Li et al., "Blueprint Separable Residual Network for Efficient Image Super-Resolution," 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), New Orleans, LA, USA, 2022, pp. 832-842, doi: 10.1109/CVPRW56347.2022.00099.
[22] Y. Mao et al., "Multi-level Dispersion Residual Network for Efficient Image Super-Resolution," 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Vancouver, BC, Canada, 2023, pp. 1660-1669, doi: 10.1109/CVPRW59228.2023.00167.
[23] C. Xie et al., "Large Kernel Distillation Network for Efficient Single Image Super-Resolution," 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Vancouver, BC, Canada, 2023, pp. 1283-1292, doi: 10.1109/CVPRW59228.2023.00135.
[24] X. Li, J. Dong, J. Tang and J. Pan, "DLGSANet: Lightweight Dynamic Local and Global Self-Attention Network for Image Super-Resolution," 2023 IEEE/CVF International Conference on Computer Vision (ICCV), Paris, France, 2023, pp. 12746-12755, doi: 10.1109/ICCV51070.2023.01175.
[25] H. Wang, Z. Wei, Q. Tang, S. Cheng, L. Wang and Y. Li, "Attention Guidance Distillation Network for Efficient Image Super-Resolution," 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Seattle, WA, USA, 2024, pp. 6287-6296, doi: 10.1109/CVPRW63382.2024.00632.
[26] D. Haase and M. Amthor, "Rethinking Depthwise Separable Convolutions: How Intra-Kernel Correlations Lead to Improved MobileNets," 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, 2020, pp. 14588-14597, doi: 10.1109/CVPR42600.2020.01461.
[27] H. Wu, L. Zhang and J. Ma, "Remote Sensing Image Super-Resolution via Saliency-Guided Feedback GANs," in IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1-16, 2022, Art no. 5600316, doi: 10.1109/TGRS.2020.3042515.
[28] Z. Wang et al., "FeNet: Feature Enhancement Network for Lightweight Remote-Sensing Image Super-Resolution," in IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1-12, 2022, Art no. 5622112, doi: 10.1109/TGRS.2022.3168787.
[29] H. Wu, N. Ni and L. Zhang, "Learning Dynamic Scale Awareness and Global Implicit Functions for Continuous-Scale Super-Resolution of Remote Sensing Images," in IEEE Transactions on Geoscience and Remote Sensing, vol. 61, pp. 1-15, 2023, Art no. 5602315, doi: 10.1109/TGRS.2023.3240254.
[30] H. Wu, N. Ni and L. Zhang, "Lightweight Stepless Super-Resolution of Remote Sensing Images via Saliency-Aware Dynamic Routing Strategy," in IEEE Transactions on Geoscience and Remote Sensing, vol. 61, pp. 1-17, 2023, Art no. 5601717, doi: 10.1109/TGRS.2023.3236624.
[31] Y. Wang, H. Zhang, X. Zeng, B. Wang, W. Li and W. Ding, "Binary Lightweight Neural Networks for Arbitrary Scale Super-Resolution of Remote Sensing Images," in IEEE Transactions on Geoscience and Remote Sensing, vol. 63, pp. 1-16, 2025, Art no. 5609716, doi: 10.1109/TGRS.2025.3529696.
[32] J. Cao et al., "CiaoSR: Continuous Implicit Attention-in-Attention Network for Arbitrary-Scale Image Super-Resolution," 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, BC, Canada, 2023, pp. 1796-1807, doi: 10.1109/CVPR52729.2023.00179.